Selling what isn't there. How online booking rewrites a pharmacy's stock and assortment

Pharmacy e-commerce in Russia is measured in several different ways. DSM Group and AlphaRM, working independently of one another, both put it at about 14.6% of medicine sales by value for 2025. Counting all pharmacy sales, parapharmaceuticals included, DSM arrives at 15.9% for 2025 and already 17.2% for the first half of 2026. RNC Pharma, for the same period, says 28.2%; Data Insight says 22–24%.

The gap between those estimates is not in the arithmetic but in what each agency counts as online. The conservative figures cover medicines only. The generous ones add parapharmaceuticals and supplements, and booking for collection in a pharmacy on top of that.

The estimates themselves bear this out. On RNC Pharma's numbers, couriers carried 15.8 billion roubles' worth of medicines in 2024, which is 0.96% of retail. In the first half of 2025 it was 10.3 billion, 1.14%. Over three quarters of 2025 delivery grew 48% in roubles and 3.5% in packs: what travels is dearer, not more frequent.

All the rest is booking goods at a pharmacy. The customer picks a medicine on a website or in an app and collects it from a pharmacy near home or work.

I advise pharmacy chains, take part in developing pricing services for retail, and am affiliated with parso.ru and flexpricer.ru.

Why the customer books rather than orders delivery

The largest pharmacy aggregator does not deliver at all. Apteka.ru says so plainly on its delivery page: "Apteka.ru does not sell goods and does not deliver orders to the home." The same page explains why: "Distance selling of medicines (including home delivery) may, under a Government Decree, be carried out only by a pharmacy organisation…" Megapteka.ru is the same — booking and collection only.

Where courier delivery does exist, it is paid for. Eapteka.ru charges from 200 roubles within Moscow. Zdravcity.ru's two-to-six-hour express costs 299. Collection is free: Eapteka.ru offers pick-up from 4,949 pharmacies at no extra charge. Against an average pharmacy basket of about 785 roubles (DSM Group, 2025), delivery eats a quarter of the sale.

Prescription medicines cannot be sold at a distance. The pilot ran for three years in Moscow, the Moscow region and the Belgorod region, and 18 pharmacy organisations were licensed for it. Sergei Glagolev, deputy health minister, gave the outcome in January 2026: some 9,700 orders in three years, more than 90% of them placed by Muscovites, about a quarter of them delivered. The pilot expired on March 1, 2026, and no extension has been announced.

Speed is less of a problem than it is assumed to be. An order placed with Eapteka.ru between 05:00 and 19:00 arrives in Moscow within three hours; Zdravcity.ru's express fits into two to six, and orders are taken around the clock. The inconvenience starts outside that window. Order in the evening and delivery moves to the next day, into slots that are generously wide. Eapteka.ru's list includes "10:00–18:00" — eight hours of waiting for a courier.

For a medicine needed today it is easier to walk to the pharmacy in your own building than to pay 200–300 roubles and wait. And if the pharmacy does not have it, the pack will travel to the pharmacy first in any case, not to the customer's door.

The result shows up in DSM Group's figures on online penetration by category. Supplements are 33.3% online, medicines 14.6%. The difference is twofold, and what creates it is regulation, not the customer's habits.

Who stands behind the pharmacy aggregators

Start with the fact that the country's three largest distributors run storefronts of their own.

These are exactly the three companies that divide the wholesale market between them. For the first half of 2026 FK Puls reported 236.65 billion roubles, Protek 231.27 billion and Katren 201.95 billion, while concentration in the distribution segment holds at 81% on DSM Group's numbers.

The storefronts are of very different sizes. More than 47,000 pharmacies are connected to Apteka.ru, which took a third of the booking segment in 2025 and turned over 83.1 billion roubles in the first half of 2026 (DSM Group). Puls launched Polza.ru in 2022 and claims more than 38,000 collection points. Protek's retail arm has been called Rigla-Zdravsiti since June 2026.

The other large aggregators have nothing to do with distribution. Uteka.ru belongs to four private individuals; Eapteka.ru has belonged since July 2026 to a joint company of Wildberries and Russ; Megapteka.ru is run by Platfomni LLC of Perm. The same pharmacy, meanwhile, usually appears on several storefronts at once.

What the price monitoring showed

Two quite different offers sit side by side on an aggregator's storefront. The first is a product that is in the pharmacy right now. The second is a product that is not, but which will be brought in tomorrow or later. The aggregators distinguish between them: Megapteka.ru's filter is labelled exactly that, "in stock" and "from the warehouse". The customer barely sees the difference; he looks at the price and the date.

I checked the proportion between the two through the price monitoring service parso.ru. In September 2026 I collected every pharmacy offer on the storefront of one of the largest aggregators across Moscow and St Petersburg. I deliberately do not name the aggregator: the subject is not that company but the design of the channel, which is the same at every large storefront. One offer is one medicine in one pharmacy, with a price and an availability flag. The totals: 1,346,825 offers from 3,712 pharmacies in Moscow, and 604,824 from 1,625 in St Petersburg.

The offers fell into two groups according to the wording on the button. "Book" means the medicine is in the pharmacy. "Order for tomorrow" or "Collect tomorrow from 10:00" means it will be brought there. This is what it comes to per pharmacy.

One item in five on a pharmacy's storefront turns out to be a product its own stock does not contain.

Where the line runs can be seen in two over-the-counter medicines whose different pack sizes diverge severalfold in availability.

Trekrezan in packs of ten is in almost every Moscow pharmacy. The same medicine at the same strength in packs of twenty is in one pharmacy in forty. Malvacid's two tablet packs are stocked by roughly equal numbers of pharmacies, the suspension by one in fourteen. The decision is neither medical nor logistical: the pharmacy is not prepared to freeze money in an item that is rarely asked for.

How the order for tomorrow works

The order for tomorrow is not filled by the pharmacy the customer collects it from. The goods travel there from a distributor's warehouse, the pharmacy takes the delivery in and hands it over.

This is how the largest aggregator describes its own business. From the same page of Apteka.ru: "An 'order' on Apteka.ru means the creation by a user of the site or mobile application of a request addressed to the supplier (Katren NPK JSC) on behalf of a pharmacy organisation for the delivery of the selected goods…" The same text says the site owner merely ensures that the goods shown on it are present in the pharmacy's assortment.

Read that again, because this is where the substance of it lies. From his own telephone the customer sends the distributor a request on the pharmacy's behalf. The distributor brings the goods to the pharmacy. The pharmacy sells them to the customer.

Zdravcity.ru describes the mechanics in more detail still. Its agreement with pharmacies states that the company creates a preliminary order and carries to the pharmacy the customer has chosen goods belonging to the company itself, which the pharmacy then receives under a supply contract. The company here is Protek — the owner of the aggregator and the distributor in one.

These two documents describe not a special case but the main mechanism of the channel. In the Moscow monitoring, 93% of all "from the warehouse" offers came from three storefronts: Zdravcity.ru 38%, Polza.ru 29%, Apteka.ru 27%. In St Petersburg the same three give 92%. Almost everything a customer orders that does not come out of a pharmacy's own stock reaches him through the storefront of one of the three largest distributors.

The main consequence is simple. A pharmacy's assortment stops being limited by its own shelf. The ceiling now sits where the distributor's warehouse ends, and that is an order of magnitude larger than any pharmacy matrix.

Then comes the part worth doing the arithmetic on. In this arrangement the pharmacy provides the premises, the licence, the pharmacist and the till, and all of that costs money every day regardless of how many orders arrive from the storefront. In return the pharmacy gets a sale through its own till and a customer who came in for the sake of that order. At the counter he can be offered something else.

Data Insight, in its review of ePharma market trends for 2025–2026, sets out the economic result in a line of its own: "The growing share of online orders turns pharmacies into collection points. Pharmacy profit in this case is formed out of manufacturers' bonuses (for large chains) and add-on sales."

On manufacturers' bonuses the industry has figures, and I give them below. Add-on sales, though, Data Insight names as a source of profit without any numbers at all, and nobody has measurements of their own. Neither the analysts nor the chains publish what share of online orders ends in an add-on sale. There is one oblique detail that proves nothing but explains where the question comes from: in the public descriptions of pharmacy inventory systems, add-on selling and online-order handling live as two separate modules, and no connection between them is described anywhere.

The prices the customer sees on an aggregator are set by the pharmacies themselves. The seller under the contract is the partner pharmacy, and it is the pharmacy that puts the price on the storefront: so it is written in the user agreements of Uteka.ru and Megapteka.ru.

What the pharmacy does give up in this arrangement is control of its own assortment. What to order from the distributor is decided for it by the customer who picked an item off the storefront. Nor does the pharmacy acquire that customer: next time he will open the website of an internet pharmacy or an aggregator again, rather than come to it.

Why pharmacies go where they earn less

If the pharmacy loses both its assortment and its customer in this process, there is no reason to agree to it. But it has almost no choice.

More than half the country's pharmacies are on the storefront already. A pharmacy that is not there the customer does not see: he is looking for a medicine, not a sign over a door, and he chooses from the list he is shown.

What the channel means for an individual chain can be seen in the accounts. The 36.6 pharmacy chain disclosed online's share of its 2025 revenue — 26.2%, more than 30 billion roubles. Like-for-like footfall in its pharmacies fell 2.9% over the same year, while the average basket rose 11.1%. More money, fewer live visitors.

The character of the competition changes too. On the shop floor a pharmacy's price is compared with one or two neighbouring signs. In an aggregator's results it stands in a common list sorted upwards, next to the prices of every pharmacy in the district that has the medicine. A fifteen-rouble difference nobody would have noticed at the counter decides here whether the order reaches the pharmacy at all.

Profit the same accounts show differently when read by legal entity. Rigla LLC grew 2025 revenue by 33%, to 93.8 billion roubles, while net profit fell 75%, from 2.3 billion to 561.4 million. Semeynaya Apteka Aprel LLC, on revenue of 52.7 billion, reported a loss of 1.1 billion. None of this can be tied to online directly. These are figures for whole legal entities, not for sales channels. Chains were also servicing debt through 2025 at a key rate that held between 21% and 16%, and digesting acquisitions of other chains. But the picture of revenue rising while profit falls is visible without any analysis at all.

What the profit is being replaced with was discussed at a February conference. A representative of the manufacturer Tonus put manufacturers' marketing payments to the chains at 20–40% of turnover. Sergei Kovalev, deputy chief executive for development at Rigla, answered that if those payments make up two thirds of profit at some chain, then that company has serious problems with its front margin. The order of this money is visible in ASNA's published materials. In its cooperation models the marketing budget runs from 3.4% to 9.5% of purchases, and the commission on its own booking service is 1.7–2.5%.

Online, the pharmacy is left with two levers: what it shows on the storefront and at what price. Everything else is decided for it by the customer and by the order-handling process.

What it costs to hold an item in stock

The pharmacy has its own version of the same model, and only one thing differs in it: whose storefront brought the order. If the customer came through a distributor's storefront, that storefront sets the price and the mark-up. If the pharmacy takes such an order itself and orders the item from its own supplier, with whom it has a contract and its own buying price, the mark-up stays with the pharmacy.

The median Moscow pharmacy does not hold in stock 78 of the items the storefront shows against its name. The median price of a product in the monitoring is 814 roubles, which at a 25% mark-up makes the buying price of such a pack 651 roubles. Take a chain of 50 pharmacies. To cover those 78 items out of its own stock it needs 3,900 packs, and at buying prices that is 2,539,000 roubles lying on the shelf waiting for a customer.

Short-term lending to small and medium-sized business cost about 18% a year by the middle of 2026, on the Bank of Russia's statistics. So 2.54 million roubles of stock costs the chain roughly 457,000 roubles a year. This is not an abstract cost of capital but real money: either interest on a loan, or the profit the same 2.54 million would have earned in circulation.

If each of those 78 items sells once a year at a 25% mark-up, the chain earns 50 × 78 × 163 roubles, or 636,000 roubles of gross profit. Holding the stock eats 457,000 of that. What is left is 179,000 roubles across 50 pharmacies, or 3,600 roubles per store per year. Net of write-offs at expiry and the cost of the shelf space, what is left is nil or less.

That is why pharmacies do not hold such items. In money terms the decision is correct, and for the customer it is an ordinary stock-out: he came for a particular pack, heard "not in stock" and went to look elsewhere.

To order, the same item costs less, because no money is frozen at all: the pharmacy buys the pack after the customer has chosen it. In place of the cost of holding stock come the aggregator's commission and the risk of a no-show.

I counted one pack per item and at the median price, so this is a floor. Real stock holds more than one pack of an item, and the dearer the item, the more money it ties up and the higher the risk to the chain.

No-shows decide whether the model works

On a booking out of stock, a no-show costs the pharmacy almost nothing. The goods were on the shelf and will stay there a while longer.

On an order placed for a particular customer the price of the error is different. The pharmacy bought a pack for one named person and that person did not come. What is left on the shelf is an item the pharmacy had deliberately chosen not to hold.

The threshold depends on whether that pack sells later. If it sells within a year or eighteen months, the error costs only the holding money, 90–180 roubles, and the model tolerates no-shows up to about 29%. If the pack goes to write-off, the error costs the full buying price and the threshold falls to 13%. I advise planning from the lower bound: no-shows above a tenth of orders swallow the whole gain.

That the aggregators know the problem themselves is visible in their terms. In September 2026 ASNA published the tariff of its booking platform Asna.ru: 1.7% on the VED list of price-capped essential medicines, 2.5% on other medicines and other goods, no subscription fee. The commission is charged on the value of orders collected, not orders placed. The aggregator does not take the no-show risk.

ASNA publishes its terms openly, which is rare in this segment. One important detail: Asna.ru is ASNA's own channel for its partner pharmacies rather than an aggregator on the model of Apteka.ru, and orders there are assembled out of the pharmacy's own stock. The rates of Apteka.ru, Zdravcity.ru, Polza.ru, Eapteka.ru, Megapteka.ru and Uteka.ru are not disclosed publicly: there are no confirmed figures for 2025–2026 in interviews, in decisions of the competition authority or in court records. Market participants know these rates, of course — each its own, out of its own contract. What the industry has no access to is a public benchmark to compare its terms against.

Online prices live by different rules

How wide the price spread is inside a single city was shown by a second round of monitoring, which I ran a week later. The median Moscow medicine sold at prices that differed by a factor of two. For 98% of items the dearest offer was more than one and a half times the cheapest, and for 56% more than twice. In St Petersburg the spread is narrower: a median of 1.8 times, with 28% of items diverging twofold or more.

These are ordinary over-the-counter medicines with more than two and a half thousand offers apiece in the city, so the extremes do not rest on one stray line.

An item off the shelf and an item to order have different costs. The first carries the cost of stock and the risk of write-off, the second the distributor's price today and the risk of a no-show. One price for both means that in one case the pharmacy is under-earning its margin and in the other it is handing the order to a competitor.

The familiar claim of "plus five to eight points of margin" has been going round presentations for a decade and has never once been confirmed against a control group. The only public Russian experiment with a control group I know of (more than 150 pharmacies, 15 against 15, two months) produced a gain of about 0.7 of a percentage point. A chain with no working category management may win 1.5–2 points of front margin in the first year; a chain whose category management is in order will win fractions of a point.

The VED list imposes a limit of its own. On Platforma OFD's data for the third quarter of 2025, price-capped essential medicines account for half the pharmacy assortment by value, by packs and by number of items alike. The pharmacy can manage price on roughly half its shelf.

That leaves the question of what exactly to manage. The price in a sorted list changes every day, so it has to be revised at the same rhythm, and by hand that is not realistic. A working arrangement is built out of three sources.

The first is daily price monitoring across your own city: what stands next to your offer in the results and where you are in the list. The second is anonymised data from fiscal data operators, which shows demand around a particular store rather than your own receipts alone. The third is data from the pharmaceutical track-and-trace system, and here I would not rush. Packs do move through it, but what the system holds and who may see it are set by law and by its operator, CRPT, rather than by what a pharmacy needs: a pharmacy works there with its own documents and sees nothing of the market around it. No service exists today that would turn track-and-trace into turnover by item and into a measure of unmet demand. If one appears, on anonymised data and with clear rules of access, it will close both holes at once, the assortment one and the pricing one. For now this is something to wait for, not a source anything can be calculated on.

The price is calculated on top of those three sources. I went through the mechanics in detail in a piece on dynamic pricing; here I will only say that the order of work is the reverse of the usual one: assortment first, price second. Putting prices in order across a matrix where a third of the items are dead weight is pointless.

What the owner of a pharmacy chain should do

This is the order in which I work through it with pharmacies and pharmacy chains.

Match the assortment on the storefront against actual stock. Export stock by store from your inventory system and compare it with what the aggregator shows against your pharmacies. The difference is the assortment you are already trading through somebody else's hands. If there is no difference, the channel is working for you as a shop window for stock you already hold, and half its point is going unused.

Calculate turnover item by item, not by drug group. The group average hides the thing that matters: inside one category some items go in a week and others stand for a year. The counting will have to be done on exports from your own inventory system: as I said above, no ready-made calculation on track-and-trace data exists yet. On the track-and-trace system itself and what it costs I have written separately, in a piece on medical device labelling.

Split the matrix into three. Items held permanently, items to order, and items that belong in neither group. The line between the first two runs through frequency of demand and the price of the pack: an expensive item with rare demand is the first candidate for ordering, a cheap one with rare demand will repay neither the holding cost nor the trouble.

Look at your prices through the customer's eyes. Open an aggregator for your city and find twenty of your fast-moving items in the results. What matters is not the average across the chain but your place in the list for a particular medicine in a particular district. The exercise takes an hour and is usually unpleasant.

Revise the online price separately from the shelf price, and more often. If competitors in your city change prices weekly and you do it monthly, for three weeks in four you are standing in the wrong place. In both directions: part of the time too dear to be ordered, part of the time too cheap for customers who would have ordered anyway.

Look at demand around the store, not only at your own receipts. Your till shows what people bought from you, not what they came for and left without, and not what is being bought in the next building.

Collect four measures on the channel. The share of aggregator orders in receipts and in revenue. The average basket on bookings against the basket on the shop floor. The no-show rate. The actual time from order to collection. Without them any conversation about the economics of the channel stays a conversation about other people's slide decks.

Check your contract with the aggregator. Who sets the price of the goods, what the commission is and what base it is charged on, and whether the aggregator can change your price without your consent. On that last question the platform economy act answers from October 1, 2026: Federal Law No. 289-FZ of July 31, 2025 bars an operator from cutting a price at the seller's expense without the seller's consent and requires non-discriminatory ranking. Whether it applies to pharmacy aggregators has not been examined publicly.

Negotiate terms of business with your distributor, not just the buying discount. Delivery times and frequency at the large distributors are much the same for every client, whereas payment deferral, return terms and order limits are negotiated individually. For the order-to-customer model these matter more than half a point on the buying price: they are what decides the cost of a pack nobody collects.

The legal frame

It is generally held that booking with collection and payment in the pharmacy needs no Roszdravnadzor licence for distance selling. No rule exists that says so in as many words. The conclusion is assembled from three separate grounds. I set them out so that you can show the construction to your own lawyer together with the aggregator's contract, rather than rely on it because I say so.

The first ground is a clarification from the Health Ministry, letter No. 25-4/433 of January 18, 2024. It is not a normative act but the ministry's position: trade is treated as distance selling when the pharmacy performs the whole set of actions, including storing the order, delivering it to the customer and maintaining a courier service. Booking includes no delivery.

The second is Article 497 of the Civil Code. A sale counts as distance selling when the customer becomes acquainted with the goods by a means that rules out direct examination at the moment the contract is concluded. Collecting a booking, the customer sees the goods and concludes the contract in the pharmacy.

The third is the way the aggregators themselves document it. In its user agreement Megapteka.ru calls an order a request to set goods aside, and the contract of sale is concluded by the customer with the partner pharmacy rather than with the service.

Prescription medicines are a separate matter, and here I advise the greatest caution. Distance selling of prescription medicines is prohibited, the pilot has ended and has not been extended. Formally a booking is not a sale, the medicine is dispensed in the pharmacy and against a prescription, and I have found no direct prohibition on setting such goods aside. But neither have I found a clear statement from the regulator that it is permitted. I would not build a channel on that without a written opinion from a lawyer.

The direction of travel in regulation is visible in a recent draft. On September 10, 2026 the Health Ministry put out for public consultation a bill making it easier for pharmacies to enter distance selling. It expressly keeps the ban on selling prescription medicines over the internet.

The state does not, for now, intend to widen prescription e-commerce, and is solving the access problem another way. In April 2026 the State Duma passed a law on a pilot scheme for mobile pharmacy points serving hard-to-reach settlements, with the procedure approved by Government Decree No. 1068 of August 24, 2026. How the economics of such a pharmacy work out I went through in a separate piece.

Questions and answers

How does booking differ from delivery? With booking the customer places an order on a website and collects and pays for it in the pharmacy. The transaction happens on the shop floor. With delivery the medicine is carried to the customer's home, and that is distance selling, with its own requirements on the seller.

What share of pharmacy sales is online? For 2025 DSM Group and AlphaRM independently give about 14.6% for medicines by value. Across all pharmacy sales, parapharmaceuticals included, DSM gives 15.9% for 2025 and 17.2% for the first half of 2026. RNC Pharma puts the same half-year at 28.2%, counting parapharmaceuticals and booking. Data Insight says 22–24%.

Does a pharmacy need a Roszdravnadzor licence to take bookings? The licence is issued for distance selling, the mark of which is delivery of the order to the customer. Booking includes no delivery. There is no direct rule on the point, so the construction is worth checking with a lawyer alongside the aggregator's contract.

Who sets the price the customer sees on an aggregator? The pharmacy itself. In the user agreements of Uteka.ru and Megapteka.ru the seller is the partner pharmacy, and it is the pharmacy that posts the price. The aggregator does not dictate it, but it does display it next to the prices of every other pharmacy in the district.

How much does a pharmacy pay an aggregator? The rates are not disclosed. The only public tariff in the segment is ASNA's: 1.7% on price-capped essential medicines and 2.5% on everything else, charged on the value of orders collected.

Can a pharmacy sell through an aggregator goods it does not hold in stock? Yes, and a substantial part of the channel is built on it. It makes economic sense for items with rare but steady demand and a high pack price. The main risk is the no-show.

Method. The full list of sources is at the end of the article. The figures from DSM Group, AlphaRM, RNC Pharma and Data Insight are not comparable with one another: each analyst has its own panel and its own denominator. The data on the presence of items in pharmacy stock and on price spreads were obtained by the author through the parso.ru retail price monitoring service across Moscow and St Petersburg, September 2026; this is a snapshot of particular days across a list of fast-moving items, and in another city or another period the figures would differ.

This material is informational and analytical in character. The medicines named are examples illustrating statistics on availability and price spread, not a recommendation to buy. The material is not advertising for them, nor medical advice, nor legal advice, including the section "The legal frame": the conclusions set out there are built on rules among which there is no direct rule on booking, and they must be checked by a lawyer against a particular contract with an aggregator. Legal statuses and quotations from aggregators' documents are as of September 22, 2026; aggregators are free to change them unilaterally.